@voldemark, very sorry it seems when I made comment 7 I must have missed
your comment 6, so I was unclear on why this was marked as affecting
Disco.
I've reproduced and will queue the patch for Bionic.
** Changed in: systemd (Ubuntu Disco)
Status: Incomplete => Fix Released
** Changed in: systemd (Ubuntu Bionic)
Assignee: (unassigned) => Dan Streetman (ddstreet)
** Changed in: systemd (Ubuntu Xenial)
Assignee: (unassigned) => Dan Streetman (ddstreet)
** Changed in: systemd (Ubuntu Xenial)
Status: Incomplete => In Progress
** Changed in: systemd (Ubuntu Bionic)
Status: Incomplete => In Progress
** Description changed:
+ [impact]
+
+ when a service configured to restart crashes, under certain
+ configurations, that service is not correctly restarted.
+
+ [test case]
+
+ see comment 8
+
+ [regression potential]
+
+ this changes the job mode of manager-triggered restarts to 'replace' any
+ existing queued job(s), instead of failing if there are queued job(s).
+ thus any regressions would occur when a service fails, that is
+ configured to restart on failure.
+
+ [scope]
+
+ This is needed only for Xenial and Bionic.
+
+ this is fixed with commit 1e0a54e72b73df7ee26e45e29ddda44c1dece919 which
+ is included already in Eoan.
+
+ [other info]
+
+ original description:
+ ---
+
+
Affected versions of OS and systemd:
$ cat /etc/issue
Ubuntu 16.04.6 LTS \n \l
$ systemd --version
systemd 229
+PAM +AUDIT +SELINUX +IMA +APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP
+GCRYPT +GNUTLS +ACL +XZ -LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD -IDN
Affected packages:
systemd 229-4ubuntu21.22 and previous versions.
Expected behaviour you didn't see:
Scheduling restart of failed service.
A process crashed by sigabrt and didn't restart.
Description:
The bug was reported to a systemd upstream repository:
https://github.com/systemd/systemd/issues/11456
The bug was fixed and accepted to the master branch:
https://github.com/systemd/systemd/pull/11467/files
Action:
Include this patch to Ubuntu 16.04 and other version of Ubuntu which are
supported.
** Description changed:
[impact]
when a service configured to restart crashes, under certain
configurations, that service is not correctly restarted.
[test case]
see comment 8
[regression potential]
this changes the job mode of manager-triggered restarts to 'replace' any
existing queued job(s), instead of failing if there are queued job(s).
thus any regressions would occur when a service fails, that is
configured to restart on failure.
[scope]
This is needed only for Xenial and Bionic.
- this is fixed with commit 1e0a54e72b73df7ee26e45e29ddda44c1dece919 which
+ this is fixed with commit 03ff2dc71ecb09272d728d458498b44f7f132f51 which
is included already in Eoan.
[other info]
original description:
---
-
Affected versions of OS and systemd:
$ cat /etc/issue
Ubuntu 16.04.6 LTS \n \l
$ systemd --version
systemd 229
+PAM +AUDIT +SELINUX +IMA +APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP
+GCRYPT +GNUTLS +ACL +XZ -LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD -IDN
Affected packages:
systemd 229-4ubuntu21.22 and previous versions.
Expected behaviour you didn't see:
Scheduling restart of failed service.
A process crashed by sigabrt and didn't restart.
Description:
The bug was reported to a systemd upstream repository:
https://github.com/systemd/systemd/issues/11456
The bug was fixed and accepted to the master branch:
https://github.com/systemd/systemd/pull/11467/files
Action:
Include this patch to Ubuntu 16.04 and other version of Ubuntu which are
supported.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1839290
Title:
systemd doesn't restart a service after crashes
To manage notifications about this bug go to:
https://bugs.launchpad.net/systemd/+bug/1839290/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
